The ultimate objective of our work is to gain sufficient insight into the mechanism of fragmentation of organic compounds in the mass spectrometer so that the principal features of the mass spectrum of a compound of known constitution can be predicted. With this information, it should then be possible to utilize the mass spectrum of an unknown compound for purposes of structure elucidation. In view of the extreme sensitivity of the method and the very small quantities required for mass spectral measurements, this technique has proved and will continue to prove itself enormously useful in the fields of natural products and of biologically important compounds (e.g., metabolites) present in very small quantities. In order to establish the ground rules of mass spectral fragmentation processes, techniques such as isotope labeling (deuterium, carbon-13 and oxygen-18) and ion cyclotron resonance will be used extensively with major emphasis on steroids. In addition, artificial intelligence techniques are being employed in developing fully automated computer methods for the analysis and structure deduction of mass spectral data, one of the aims being to provide a rapid screening method for naturally occurring marine sterols.